H-Infinity Disturbance-Attenuating Latent Observer / bench_report.json

Failed on benchmark

Raw ⬇ ZIP
 1{
 2  "bench_version": 1,
 3  "track": "dynamics",
 4  "model": "rnn_small",
 5  "metric_direction": "lower is better",
 6  "n_seeds": 8,
 7  "baseline": {
 8    "best_cfg": {
 9      "lr": 0.006
10    },
11    "sweep": [
12      {
13        "cfg": {
14          "lr": 0.001
15        },
16        "mean": 0.0015310924500226974
17      },
18      {
19        "cfg": {
20          "lr": 0.003
21        },
22        "mean": 0.0008559521666029468
23      },
24      {
25        "cfg": {
26          "lr": 0.006
27        },
28        "mean": 0.0005687399534508586
29      }
30    ],
31    "full": {
32      "mean": 0.0005194859513721894,
33      "std": 0.0001693891077925739,
34      "per_seed": [
35        0.0005914251087233424,
36        0.0007362987380474806,
37        0.00035295303678140044,
38        0.0005942829302512109,
39        0.0002958264376502484,
40        0.0005069610779173672,
41        0.00032333534909412265,
42        0.0007548049325123429
43      ],
44      "n": 8
45    }
46  },
47  "idea": {
48    "mean": 0.0005194859513721894,
49    "std": 0.0001693891077925739,
50    "per_seed": [
51      0.0005914251087233424,
52      0.0007362987380474806,
53      0.00035295303678140044,
54      0.0005942829302512109,
55      0.0002958264376502484,
56      0.0005069610779173672,
57      0.00032333534909412265,
58      0.0007548049325123429
59    ],
60    "n": 8
61  },
62  "comparison": {
63    "delta_mean": 0.0,
64    "idea_wins": 0,
65    "n_pairs": 8,
66    "per_seed_diffs": [
67      0.0,
68      0.0,
69      0.0,
70      0.0,
71      0.0,
72      0.0,
73      0.0,
74      0.0
75    ],
76    "p_value": 1.0,
77    "mde": 0.0,
78    "mde_rel_pct": 0.0,
79    "verdict": "no measurable effect",
80    "system_worked": false
81  },
82  "mechanism_signature": {
83    "claim": "bounded-real attenuation reduces local input-disturbance gain",
84    "gamma": 0.5,
85    "idea_lr": 0.006,
86    "baseline_observed_mean_local_gain": 0.431841354817152,
87    "idea_observed_mean_local_gain": 0.431841354817152,
88    "baseline_observed_p95_local_gain": 0.4626914465799927,
89    "idea_observed_p95_local_gain": 0.4626914465799927,
90    "baseline_task_mse_at_same_lr": 0.0005194859513721894,
91    "idea_task_mse_at_same_lr": 0.0005194859513721894,
92    "confirmed": false
93  }
94}